AI and Covid-19 Vaccination
(PRUnderground) April 23rd, 2021
Public biases in healthcare, classically caused by lack of confidence to clinical trials, have led to trust issues among some considerable portion of the populations. Research shows that the lack of trust and confidence is a serious issue even among some healthcare workers who tend to demonstrate lower vaccine adoption. Majority of the mistrust is about adverse effects. Can artificial intelligence (AI) help in convincing these group of people on safety of COVID-19 vaccine?
Honest public relationship and transparency on adverse effects of COVID-19 could be a factor to improve communication and potentially reduce hesitance among populations. One way to implement such a communication strategy is to employ AI-based technologies. There have been some attempts in tracking down the side effects of vaccine through smart phone apps. For example, CDC is using V-safe app that to employ monitoring and just-in-time indicating of the COVID-19 vaccine roll-out. V-safe and similar apps in the market have been successful in giving people the capability to provide valuable and up-to-date information of adverse effects of COVID-19 vaccine. These tools and mobile apps have been designed to assist improving the trust into COVID-19 vaccine and increasing the rate of up-take. “However, they only go half-way in their mission.” Dr. Amir Talaei-Khoei said, “They collect valuable data and do nothing with it.” We are living in the world that data is coming from everywhere but putting it together and building meaningful models that can be used in practice is the real deal.
AI-enabled tools and machine learning algorithms can be loaded with the data coming from platforms like V-safe to generate real-time up-to-date models for the adverse effects of COVID-19 vaccine. Dr. Amir Talaei-Khoei is now working on developing a tool that can generate such side-effect tools for COVID-19 vaccines. The models will represent the actual administration of the vaccination programs not only based on demographics of vaccinated individuals but also according to the geographical locations, cultural constructs and the vaccine brand.
One of the biggest issues in deploying an AI-based tool to build the model of adverse effects for COVID-19 vaccine is the limited self-reported date available to train these models. “AI models are as good as the data.”, Dr. Amir Talaei-Khoei said, “Therefore, the forecasts of the adverse effects of the vaccine can be subject to various and serious biases, which might lead to another risk for the lack of confidence and trust among population.” In order to overcome the bias in the data and also in the AI-tools, frequent but accurate audit would be required.
Such development in COVID-19 vaccination programs will help to simplify the seemingly inexorable uncertainty that has come to characterize the today’s world.
About Medical AI SystemsPress Contact
- Name: Jason Ardashiry
- Phone: 775 624 5677
- Email: Contact Us
Original Press Release.